AMDGPU: Fix verifier errors in SILowerControlFlow
The main sin this was committing was using terminator
instructions in the middle of the block, and then
not updating the block successors / predecessors.
Split the blocks up to avoid this and introduce new
pseudo instructions for branches taken with exec masking.
Also use a pseudo instead of emitting s_endpgm and erasing
it in the special case of a non-void return.
llvm-svn: 273467
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
index a580d80..2b13bb9 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
@@ -261,5 +261,8 @@
//===----------------------------------------------------------------------===//
// Call/Return DAG Nodes
//===----------------------------------------------------------------------===//
-def IL_retflag : SDNode<"AMDGPUISD::RET_FLAG", SDTNone,
+def AMDGPUendpgm : SDNode<"AMDGPUISD::ENDPGM", SDTNone,
+ [SDNPHasChain, SDNPOptInGlue]>;
+
+def AMDGPUreturn : SDNode<"AMDGPUISD::RETURN", SDTNone,
[SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;